Proteomics: Prediction and Analysis of Protein Networks

We are amongst the first groups to describe how many weak functional genomic features could be systematically integrated with data mining techniques to predict protein networks (comprising protein interactions and other functional linkages). Some of the features integrated were obviously related to protein interactions (e.g. expression correlations) but many others such as (e.g. essentiality) were less so. We have had a number of localization and interaction predictions experimentally verified. In more recent work, we were able to calibrate the degree to which the data quality and the specific mining approach is associated with the strength of the predictions. In addition, we have studied the structure of protein networks, both on a large-scale in terms of global statistics (e.g. the diameter) and on a small-scale in terms of local network motifs (e.g. hubs). In particular, we have correlated network hubs with gene essentiality. Most importantly, we were the first to study the dynamics of regulatory networks. This allowed us to discover changing transient hubs and systematic patterns of connectivity rewiring in the yeast regulatory network. We were able to show for the first time that network dramatically changes in different conditions.